Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Jul 1.
Published in final edited form as: J Med Genet. 2010 May 27;47(7):499–506. doi: 10.1136/jmg.2009.075143

Novel mutations in the long isoform of the USH2A gene in patients with Usher syndrome type II or non-syndromic retinitis pigmentosa

Terri L McGee 1,2, Babak Jian Seyedahmadi 1, Meredith O Sweeney 1, Thaddeus P Dryja 1, Eliot L Berson 2
PMCID: PMC3070405  NIHMSID: NIHMS253518  PMID: 20507924

Abstract

Background

Usher syndrome type II (USH2) is an autosomal recessive disorder characterized by retinitis pigmentosa (RP) and mild to moderate sensorineural hearing loss. Mutations in the USH2A gene are the most common cause of USH2 and are also a cause of some forms of RP without hearing loss (ie non-syndromic RP). The USH2A gene was initially identified as a transcript comprised of 21 exons but subsequently a longer isoform containing 72 exons was identified.

Methods

The 51 exons unique to the long isoform of USH2A were screened for mutations among a core set of 108 patients diagnosed with USH2 and 80 patients with non-syndromic RP who were all included in a previously reported screen of the short isoform of USH2A. For several exons, additional patients were screened.

Results

In total, 35 deleterious mutations were identified including 17 nonsense mutations, 9 frameshift mutations, 5 splice-site mutations, and 4 small in-frame deletions or insertions. Twenty-seven mutations were novel. In addition, 65 rare missense changes were identified. A method of classifying the deleterious effect of the missense changes was developed using the summed results of 4 different mutation assessment algorithms, SIFT, pMUT, PolyPhen, and AGVGD. This system classified 8 of the 65 changes as “likely deleterious” and 9 as “possibly deleterious”.

Conclusion

At least one mutation was identified in 57–63% of USH2 cases and 19–23% of cases of non-syndromic recessive RP (calculated without and including probable/possible deleterious changes) thus supporting that USH2A is the most common known cause of RP in the United States.

Keywords: Usher syndrome, retinitis pigmentosa, mutation, deafness

INTRODUCTION

Usher syndrome is an autosomal recessive condition where patients have both retinitis pigmentosa and sensorineural hearing loss and, in some cases, vestibular defects. It is the most common cause of deaf-blindness and its prevalence is estimated to be between 3.3 – 6.4 per 100,000 people.[1–5] This syndrome is divided into three clinical subtypes based on the severity and onset of the deafness and the presence or absence of vestibular defects. Usher syndrome type I (USHI) is characterized by RP, profound congenital hearing loss and vestibular dysfunction. Patients with Usher type II (USH2) have RP with partial, early onset hearing loss and no vestibular defect. Usher type III features RP, progressive hearing loss, normally of adult onset and variable vestibular dysfunction. Genetic heterogeneity is evident in Usher syndrome with USHI caused by mutations in any one of 6 genes (MYO7A,[6] USH1C,[7] CDH23,[8, 9] PCDH15,[10, 11] and USH1G[12]), and the as yet unidentified gene assigned to chromosome 21q21 (USH1E[13]), USH2 caused by three identified genes (USH2A,[14] GPR98 (formerly VLGR1 or MASS1),[15] and DFNB31(Whirlin) [16]) and at least one unidentified gene,[17] and USH3 caused by mutations in the gene CLRN1.[18] USH2 is the most common form of Usher syndrome.[1, 3, 4] However this may not be the case in all populations where for example, a founder effect may skew the distribution of cases.[5, 19, 20] Of patients with USH2, it is estimated that between 75% to 90% have pathogenic mutations in the USH2A gene[21–23] making this gene the most common known cause of Usher syndrome.

Mutations in the USH2A gene were first reported in patients with USH2 in 1998.[14] This first study analyzed the 21 exons encoding the originally described USH2A transcript.[24] A longer transcript consisting of 51 additional 3’ exons was subsequently identified and mutations causing USH2 were also found in those exons.[25] In a recent study, comprehensive analysis of the newly characterized 51 exons revealed allelic mutations in approximately half of the patients in whom one pathogenic mutation had been identified in a screen of the shorter isoform.[22] More than 70 different null alleles and several different missense mutations have been identified in the USH2A gene in patients with USH2 (for a summary see the UMD-USHbases http://www.umd.be/USH2A/[26]). Of these, most are found in only one or a few cases each. The exception to this is the common mutation c.2299delG (Glu767fs) which has been reported as a frequent mutation in several populations and found to have been derived from a common ancestor.[27] Comprehensive screens of the entire long isoform revealed that approximately 60% of the unique pathogenic mutations occur in the exons specific to the long isoform.[21, 23]

One corollary of the fact that many mutations are solitary or found only in a few cases is that inferring the pathogenicity of such sequences can not be confidently accomplished even if the change is not observed among a hundred normal control subjects. For example, a sequence variation would need to be observed in at least 6 out of 200 disease chromosomes if not found in any of 200 normal chromosomes before being considered statistically significantly associated with disease (p < 0.05 using a Fisher’s exact test). Rare nonsense, frameshift, or splicing mutations can be interpreted as pathogenic null alleles since they are highly likely to be null alleles. However, rare missense variations could be either deleterious or harmless. Computational algorithms for analyzing the consequences of missense substitutions have been developed, and they largely depend on protein sequence alignments across species.[28] In this study the interpretation of identified rare missense changes is aided by a combined analysis of 4 such computational programs.

We previously reported a comprehensive screen of the short isoform in a cohort of patients and found that 79% of the patients with identified USH2A mutations had only one mutant allele detected.[29] One purpose of this current study was to analyze the long isoform to identify additional patients with mutations in this gene and to identify the second mutation in patients with one previously detected USH2A mutation.

MATERIALS AND METHODS

Patients

This study adhered to the tenets of the Declaration of Helsinki and was approved by the Institutional Review Boards of the Massachusetts Eye and Ear Infirmary and Harvard Medical School. Informed consent was obtained from patients included in this study. Over 95% of the patients included in this study were Caucasian which reflects the patient base of the Berman-Gund Laboratory at Massachusetts Eye and Ear Infirmary. The clinical phenotypes of many of these patients have recently been described. [30]

We sequenced the coding exons specific to the long isoform of the USH2A gene (exons 22–72) in 108 patients diagnosed with Usher syndrome type II and in an additional 80 patients with non-syndromic retinitis pigmentosa. The diagnosis of Usher syndrome was made based on an ocular examination, electroretinography, and self-reported hearing loss. Hearing tests were not administered to most patients and therefore some of the patients diagnosed as having non-syndromic retinitis pigmentosa may in fact have some hearing deficit. Most (85%) of the patient samples did not yield data for at least one of the exons screened such that 4.7% of the 9776 exon fragments (188 samples × 52 amplicons) were not screened. Exons 1–21 of the USH2A gene were previously sequenced in this set of patients.[29] For select exons additional probands (beyond the 188 core set) were screened as part of the effort to identify more USH2A patients. The patients included in this study were also screened for mutations in the USH3A gene as part of a concurrent study in a larger set of patients.[31]

DNA sequencing

Oligonucleotide primers designed for each exon and 20 ng of patient DNA were used in the polymerase chain reaction to amplify each exon and nearby flanking intron sequence. Amplified DNA was treated with the enzyme ExoSAP-it (US Biochemical) to eliminate unincorporated PCR primers and dNTPs. The treated sample was then sequenced using the BigDye Terminator v3.1 cycle sequencing reaction and the ABI 3100 Genetic Analyzer (AppliedBiosystems). All sequence variants identified were confirmed by sequencing in both the sense and antisense direction.

Assessment of pathogenicity

For this study, we have classified all nonsense mutations, frameshift mutations, and mutations of the first two bases of canonical intron splice donor or acceptor sites as deleterious. Missense changes with an allele frequency >4% in patients and normal controls and isocoding changes that were not predicted to alter splicing were classified as benign. Rare missense changes were further analyzed using 4 different computer-generated algorithms (SIFT, PolyPhen, AGVGD, and pMut) designed to predict whether a variation is deleterious. The combined results of these computer analyses were used to develop a classification system for each missense change as follows. First, a score of 0 (benign), 0.5 (possibly deleterious), or 1 (probably deleterious) was assigned based on the results of each of the 4 tests. The scores for all 4 tests were then summed and missense variants with a summed score of 4.0 were classified as “likely deleterious”, those with a summed score of 3.0–3.5 were classified as “possibly deleterious”, those with a total score of 2.0–2.5 were classified as “unknown”, and those with scores <2.0 were classified as likely benign. In addition to the variants identified in the screen of exons 22–72, we used these 4 algorithms to reclassify the missense changes from exons 1–21 that were previously characterized as variants of uncertain pathogenicity by our group.[29] A brief description of the 4 algorithms used in this analysis is below.

  1. SIFT: The program, Sort Intolerant From Tolerant (SIFT) (available at http://sift.jcvi.org) relies on protein sequence alignments of multiple species and considers the conservation of the affected residue and the type of amino acid substitution to predict whether a change is innocuous or deleterious.[32, 33] It has been predicted that limiting the alignment used for the SIFT the analysis to true orthologs may limit false negatives since paralogs may have slightly different sequences that account for their subtle functional differences.[32, 34] We created a list of USH2A orthologs through the use of a BLAST search and an ortholog search in the ENSEMBL database and used CLUSTALW to create a homology alignment of the USH2A/ orthologs from human, chimpanzee, horse, dog, vampire bat, armadillo, sloth, opossum, chicken, zebra finch, platypus, zebra fish, stickleback, sea urchin, sea squirt, rat, mouse, and guinea pig. The CLUSTAL alignment was used for the SIFT analysis of the rare missense changes identified in this study and achieved the recommended level of diversity with a median conservation score of ≤3. The output from the SIFT analysis is a binary classification of “tolerated” or “affect”. For our classification, a score of 0 was given for an output of “tolerated” and a score of 1 given for an output of “affect”.

  2. PolyPhen: The polymorphism phenotyping program, PolyPhen (available at http://genetics.bwh.harvard.edu/pph/) is based on an empirical set of rules that combines analyses of a pre-built multiple protein sequence alignment with a number of protein structural and functional attributes to estimate the consequence of an amino acid substitution which is then characterized as benign, possibly deleterious, or deleterious.[35] We assigned a score of 0 to variations characterized as benign, a score of 0.5 for those characterized as possibly deleterious, and a score of 1.0 for those characterized as deleterious.

  3. Pmut: Pmut (available at http://mmb2.pcb.ub.es:8080/PMut/) is a program that uses neural networks trained on known disease-causing human mutations and neutral mutations to provide the binary prediction of “neutral” or “pathologic”.[36] Although the web portal indicates the ability to insert one’s own alignment for analysis, we as well as another group[28] could not get this option to operate successfully so the default alignment was used. A score of 0 was assigned to variants classified as neutral and a score of 1.0 was given to those classified as pathologic for our classification system.

  4. Align-GVGD: The program Align-GVGD (available at http://agvgd.iarc.fr/agvgd_input.php) combines an assessment of an alignment with the physicochemical characteristics of the amino acid to calculate the Grantham variation (GV) at each position and the distance the substituted base is from the range of variation (GD).[37] We used the same alignment of orthologs used with the SIFT program for this analysis. Based on the GV and GD measurements, a grade of C0-C65 is given where C0 is benign and C65 is most likely pathogenic. For our classification system, a score of 0 was given to grade C0; a score of 0.5 was given to grades C15, C25, or C35; and a score of 1.0 was given for a grade of C45, C55, or C65.

RESULTS

Mutation analyses

Analysis of the sequence of exons 22 through 72 identified 35 mutations interpreted as deleterious (Table 1, Fig. 1). Of these, 17 were nonsense mutations, 9 were frameshift mutations, 4 were mutations of canonical AG or GT splice acceptor or donor sites, 4 were small in-frame deletions or insertions, and one was a missense change, Glu1926Lys, predicted to alter a splice site (prediction score using splice-site prediction software http://www.fruitfly.org/seq_tools/splice.html was reduced from .54 to <.1). These mutations were deemed deleterious since they would substantially alter the encoded protein. Twenty-seven of these mutations were identified in only one patient each and the remaining 8 mutations were identified in 2–7 patients in our study. Twenty-seven of the mutations were novel.

Table 1.

Deleterious mutations identified in exons 22–72 of the USH2A gene and their distribution among patients with USH2 or recessive non-syndromic retinitis pigmentosa (ARRP).

exon Deleterious mutations identified in exons 22–72 protein change distribution among all Usher
type II patients screened		 distribution among all
ARRP patients screened
nucleotide change homoz hetero wild type homoz hetero wild type
22 c.4645C>T Arg1549X[21] 0 4 206 0 0 145
22 c.4758+1G>A n/a 0 0 210 0 1 144
26 c.5191_5192delAT Met1731ValfsX8* 0 1 217 0 0 150
27 c.5514T>A Tyr1838X[38] 0 1 194 0 0 139
28 c.5776G>A Glu1926Lys ** 0 1 216 0 0 149
28 c.5836C>T Arg1946X 0 1 216 0 0 149
int28 c.5775+1G>A n/a[23] 0 1 216 0 0 149
int29 c.5857+1G>C n/a* 0 1 216 0 0 149
int29 c.5857+2T>C n/a* 0 2 215 0 0 149
30 c.5933_5940delCTGTTGTC Pro1978GlnfsX5* 0 1 207 0 0 141
32 c.6169C>T Gln2057X* 0 1 192 0 0 130
32 c.6289_6302delATCTATTCAGGCAG Ile2097fsX1 0 1 192 0 0 130
40   c.7493delG Ser2498MetfsX30* 0 1 99 0 0 67
41 c.8167C>T Arg2723X* 1 2 101 0 0 73
42 c.8557A>T Arg2853X[23] 0 1 107 0 0 76
44   c.8740C>T Arg2914X 0 1 75 0 0 57
44 c.8834G>A Trp2945X* 0 1 75 0 0 57
47 c.9270C>A Cys3090X 0 1 104 0 0 76
47 c. 9336_9359delTATACCAACACCCACAATTCGTGG Ile3113_Gly3120del 0 1 104 0 0 76
48 c.9424G>T Gly3142X[21] 0 2 105 0 0 80
50 c.9787_9807delATTGGTGATTCCTGCTGTGGC Ile3263_Gly3269del 0 1 107 0 0 80
52 c.10190_10191delAA Lys3397ArgfsX20* 0 1 107 0 0 80
59 c.11533C>T Gln3845X* 1 0 80 0 0 53
61 c.11864G>A Trp3955X[25] 0 2 184 0 5 140
61 c.12006C>A Tyr4002X 0 1 185 0 0 145
int61 c.12067-2A>G n/a[39] 1 1 103 0 0 82
63 c.12490delC His4164ThrfsX3 0 1 203 0 0 141
63 c.12954C>A Tyr4318X 0 1 106 0 0 79
63 c.13335_13347delGAACATGGACTCTinsCTTG Glu4445_Ser4449del-insAspLeu 0 0 108 0 2 79
63 c.13499_13500 insTACTCTCAC Thr4500_Pro4501insThrLeuThr 0 1 107 0 0 80
64 c.14010_14062del Glu4671AsnfsX4* 0 0 98 0 1 78
64 c.14131C>T Gln4711X* 0 1 97 0 0 79
67 c.14787delA Glu4930AsnfsX20 0 0 104 0 1 64
68 c.14803C>T Arg4935X[21] 0 3 103 0 1 79
68 c.14879_c.14880delAAins45 Gln4960Leufs* 0 1 105 0 0 80
Open in a new tab

References for previously reported mutations are indicated in the column labeled “protein change”.

*

mutations for which the clinical findings of patients have recently been reported[30]

**

a missense change that is predicted to destroy an intron splice site (splice site probability reduced from .54 to <.1- see results) and is thus classified as deleterious, Homoz = number of patients homozygous for rare allele, Hetero = number of patients heterozygous for the rare allele, Wild-type = number of patients homozygous for the wild type, or normal sequence at the respective position. Total numbers are not equal as some samples did not yield data for some exons and some exons were screened in additional sets of patients (see methods).

Figure 1.

Figure 1

Open in a new tab

Schematic illustration of the USH2A gene and the encoded usherin protein. The location of all deleterious, probably deleterious, and possibly deleterious mutations found in our patients are indicated. The locations of amino acids altered by the missense mutations are also indicated on the schematic representation of the usherin protein.

We also identified 65 rare missense changes (<4% allele frequency) (Table 2), 17 of which have been previously reported. Most of the 65 changes were found in only one patient each and, therefore, analyzing ~100 normal controls for these mutations would be unlikely to provide a meaningful assessment of whether the change was disease-causing.

Table 2.

Classification of rare missense variations identified in exons 22–72 of USH2A and their distribution among patients with USH2 or recessive non-syndromic retinitis pigmentosa (ARRP).

Rare missense changes Protein prediction – deleterious effect Distribution among
USH2		 distribution among
ARRP
exon nucleotide
change		 protein change Sift PolyPhen pMut AGVGD Score homoz Hetero wild
type		 homoz hetero wild
type
27 c.5506C>A Pro1836Thr tolerated/.17 poss/1.951 neutral/8 C0 .5 0 1 194 0 0 139
30 c.5858C>G Ala1953Gly affect/.05 poss/1.505 neutral/8 C0 1.5 0 1 207 0 1 140
30 c.5932C>T Pro1978Ser affect/0.0 prob/2.396 neutral/7 C65 3 0 0 208 0 1 140
30 c.5975A>G Tyr1992Cys affect/0.0 prob/2.758 path/6 C35 3.5 0 2 206 0 0 141
32 c.6240G>T Lys2080Asnb[23] tolerated/.1 benign neutral/7 C0 0 0 1 192 0 1 129
33 c.6347A>G His2116Arg tolerated/.06 prob/2.413 neutral/3 C0 1 0 1 180 0 0 83
33 c.6383G>A Cys2128Tyr affect/0.0 prob/3.346 path/8 C65 4 0 1 180 0 0 83
33 c.6383G>T Cys2128Phe affect/0.0 prob/3.346 path/2 C65 4 0 1 180 0 0 83
34 c.6587G>C Ser2196Thr tolerated/.74 benign/.192 neutral/8 C0 0 0 2 202 0 2 136
35 c.6709G>T Asp2237Tyr affect/.01 prob/2.104 path/3 C15 3.5 0 0 108 0 1 77
35 c.6713A>C Glu2238Alab[22] affect/.01 prob/2.059 neutral/7 C15 2.5 0 3 105 0 1 77
35 c.6778T>C Ser2260Pro affect/.01 poss/1.729 neutral/1 C0 1.5 0 1 107 0 0 78
36 c.6875 G>A Arg2292Hisb[23] tolerated/.65 prob/2.312 neutral/1 C0 1 0 7 176 0 3 79
41 c.7685T>C Val2562Alab[23] tolerated/.26 benign/.496 neutral/7 C0 0 0 1 102 0 4 69
41 c.7718G>A Arg2573His tolerated/.24 benign/.95 neutral/2 C0 0 0 0 104 0 1 72
41 c.7915T>C Ser2639Pro affect/.01 poss/1.71 neutral/0 C0 1.5 0 2 102 0 0 73
42 c.8357T>C Phe2786Ser affect/0.0 prob/2.208 path/1 C55 4†‡ 0 2 106 0 0 76
43 c.8624G>A Arg2875Glnb[22] affect/.03 benign/0.0 neutral/1 C0 1 0 4 89 0 6 69
43 c.8656C>T Leu2886Pheb[22] tolerated/1.0 poss/1.96 neutral/7 C0 0.5 0 8 85 0 4 71
44 c.8790T>G Asn2930Lys tolerated/.13 benign/1.41 neutral/4 C0 0 0 0 76 0 2 55
47 c.9262G>A Glu3088Lysb[23] affect/0.0 poss/1.609 neutral/4 C55 2.5 0 2 103 0 1 75
47 c.9343A>G Thr3115Alab[23] tolerated/1.0 benign/.238 neutral/7 C0 0 0 9 96 0 4 72
49 c.9595A>G Asn3199Aspb[23] tolerated/1.0 benign/.421 neutral/9 C0 0 0 9 94 0 3 70
50 c.9799T>C Cys3267Arga[22] affect/0.0 prob/3.571 path/5 C65 4 0 1 107 0 0 80
51 c.10073G>A Cys3358Tyr affect/0.0 prob/3.346 path/8 C65 4 0 0 84 0 2 74
51 c.10151C>A Ser3384Tyr affect/.00 poss/1.954 path/5 C15 3 0 0 84 0 1 75
52 c.10342G>A Glu3448Lys tolerated/.21 poss/1.52 neutral/5 C0 0.5 0 1 107 0 0 80
52 c.10385C>T Thr3462Ile tolerated/.59 benign/.22 path/1 C0 1 0 1 107 0 0 80
53 c.10437G>T Trp3479Cys tolerated/.07 prob/3.308 path/5 C15 2.5 0 1 103 0 0 78
53 c.10561T>C Trp3521Arga[23] affect/0.0 prob/3.902 path/8 C65 4 0 3 101 0 0 78
53 c.10575G>A Gly3529Ser affect/0.0 poss/1.824 neutral/5 C55 2.5 0 1 103 0 0 78
55 c.10817T>C Leu3606Pro affect/.0 prob/2.076 path/1 C25 3.5 0 0 99 0 1 72
55 c.10852G>A Gly3618Ser affect/.03 poss/1.555 neutral/6 C0 0 0 0 99 0 1 72
57 c.11156G>A Arg3719His affect/.0 poss/1.979 neutral/3 C25 2 0 0 99 0 1 70
59 c.11504C>T Thr3835Ileb[23] tolerated/.1 poss/1.567 path/0 C0 0 5 1 74 1 2 50
59 c.11532A>G Ile3844Met affect/.01 benign/1.425 neutral/9 C0 1 1 0 80 0 0 53
60 c.11677C>A Pro3893Thrb[23] affect/0.0 prob/2.396 neutral/6 C35 2.5 0 5 102 1 2 77
60 c.11711G>A Arg3904Lys affect/0.0 poss/1.529 neutral/9 C25 2 0 1 106 0 0 80
62 c.12282T>G Asn4094Lys affect/0.0 prob/2.063 neutral/6 C65 3 0 0 105 0 1 81
63 c.12522C>G Ser4174Arg tolerated/.53 benign/.063 path/0 C0 1 0 1 203 0 0 141
63 c.12575G>A Arg4192His tolerated/.23 benign/.424 neutral/2 C0 0 0 0 204 1 1 139
63 c.12742C>A His4248Asn tolerated/ prob/2.073 neutral/8 C0 1 0 0 204 0 1 140
63 c.12806C>G Pro4269Arg affect/0.0 prob/2.621 path0 C65 3 1 0 106 0 0 79
63 c.13010C>T Thr4337Meta[22] affect/0.0 poss/1.835 neutral/1 C15 2 0 1 106 0 0 79
63 c.13297G>T Val4433Leu tolerated/.30 benign/.969 neutral/6 C0 0 0 3 104 0 2 77
63 c.13339A>G Met4447Val affect/.02 prob/2.266 neutral/5 C0 2 0 0 108 0 1 79
63 c.13709G>A Arg4570His tolerated/.20 benign/.32 neutral/0 C0 0 0 1 107 0 1 79
64 c.13984C>G Gln4662Glu Tolerated/.36 poss/1.546 neutral/9   C0 0.5 0 1 97 0 0 79
64 c.14074G>A Gly4692Arg affect/.02 poss/1.913 path/2 C0 2.5 0 1 97 0 0 79
65 c.14287G>A Gly4763Arg affect/.00 prob/2.274 path/3 C65 4 0 1 107 0 1 79
65 c.14333C>A Ala4778Asp tolerated/.07 benign/1.237 neutral/0 C0 0 0 1 107 0 0 80
66 c.14422T>C Cys4808Arg affect/0.0 prob/3.571 path/6 C65 4 0 1 137 0 0 114
66 c.14449G>A Gly4817Arg affect/.00 poss/1.198 path/3 C25 3.5 0 1 137 0 0 114
66 c.14453C>T Pro4818Leua[22] affect/.02 prob/2.567 path/5 C0 3 0 1 137 0 0 114
66 c.14513G>A Gly4838Glu tolerated/1.0 benign/.773 path/2 C0 0 0 0 138 0 1 113
66 c.14519T>C Leu4840Pro affect/02 poss-1.58 path/0 C0 2.5 0 0 138 0 1 113
66 c.14531C>T Thr4844Met tolerated/.08 benign/.34 neutral/1 C0 0 0 0 138 0 1 113
66 c.14543G>A Arg4848Gln tolerated/1.0 benign/.396 neutral/3 C0 0 0 0 138 0 2 112
67 c.14753C>T Thr4918Met affect/.01 benign/1.421 path/1 C0 2 0 1 103 0 0 65
70 c.15076A>G Lys5026Glu tolerated/1.0 benign/1.382 neutral/4 C0 0 0 0 108 0 2 78
70 c.15091C>T Arg5031Trpb[23] tolerated/.07 benign/.298 path/9 C0 1 0 1 107 1 3 76
71 c.15377T>C Ile5126Thr[38] tolerated/.09 benign/1.434 neutral/5 C0 0 0 3 100 0 1 80
71 c.15428G>A Arg5143His affect/.03 poss/1.707 neutral/1 C0 1.5 0 0 103 0 2 79
71 c.15433G>A Val5145Ile tolerated/.58 benign/.992 neutral/9 C0 0 0 0 103 0 1 80
72 c.15562A>G Ser5188Gly tolerated/.46 poss/1.636 neutral/7 C0 0.5 0 0 107 0 2 76
Open in a new tab

The output from 4 different sequence analysis programs used to predict pathogenicity are shown along with a summed score. Variations with a score of 4.0 were classified as “likely deleterious”, those with a score of 3.0–3.5 were classified as “possibly deleterious”, those with a score of 2.0–2.5 were classified as “unknown”, and those with a score <2.0 were classified as “benign”. References of previously published changes are indicated in the column labeled “protein change”.

a

previously reported as pathogenic

b

previously reported as non-pathogenic

†

patient with this change already has two other clearly pathogenic mutations in USH2A

‡

patient with this change has a deleterious mutation in USH3A gene, Homoz = number of patients homozygous for rare allele, Hetero = number of patients heterozygous for the rare allele, Wild-type = number of patients homozygous for the wild type sequence at the respective position. Total numbers are not equal as some samples did not yield data for some exons and some exons were screened in additional sets of patients (see methods).

To help assess the possible deleterious effect of the rare missense changes, each was evaluated with the 4 different computer programs SIFT, PolyPhen, Pmut, and AGVGD. The outputs of these individual analyses are summarized in table 2 along with their summed scores (as described in the methods section) which were used to classify the variations. Using this scoring system, 8 missense changes (Cys2128Tyr, Cys2128Phe, Phe2786Ser, Cys3267Arg, Cys3358Tyr, Trp3521Arg, Gly4763Arg, and Cys4808Arg) had scores of 4.0 and were thus initially classified as “likely deleterious”. Two of these changes were previously reported as pathogenic variants (Cys3267Arg and Trp3521Arg).[22, 23] The change Phe2786Ser was found in one patient who was later found to have a mutation in the USH3A gene and one patient who had two other clearly deleterious mutations in the USH2A gene, so it is unclear whether the variant Phe2786Ser contributes to Usher syndrome in these 2 patients. Nine missense changes (Pro1978Ser, Tyr1992Cys, Asp2237Tyr, Ser3384Tyr, Leu3606Pro, Asn4094Lys, Pro4269Arg, Pro4818Leu, and Leu4840Pro) had scores of 3.0–3.5 and were therefore classified as “possibly deleterious”. One of the “possibly deleterious” variants (Pro4818Leu) was previously described as pathogenic.[22] The deleterious effects of 12 missense changes were classified as “unknown” with scores of 2.0–2.5. One of these changes (Thr4337Met) was previously reported to be pathogenic[22] and 3 (Glu2238Ala, Glu3088Lys, and Pro3893Thr) were reported previously as nonpathogenic.[22, 23] The remaining 36 rare missense variations had scores <2.0 and were classified as “likely benign”. One of these, Ile5126Thr, was recently reported as pathogenic[38] yet had a score of 0.0 using our rating system. The location of the missense changes found to be likely or possibly deleterious is illustrated in Fig. 1.

This same pathogenicity analysis was performed on the 11 missense variations previously reported by our group as being of uncertain pathogenicity.[29] Using this scoring system, none of the missense mutations from the screen of exons 1–21 could be classified as “likely deleterious” (i.e., all scores were < 4.0). The variants Arg464Cys and Arg517Thr were classified as “possibly deleterious” with a score of 3.0; the variants Ser307Ile, Ser391Ile, Gly516Val, Cys575Ser, and Pro1059Leu were classified as “unknown”; and Glu478Asp, Phe739Leu, Thr911Asn, and Leu1470Arg were classified as “likely benign”.

Finally, 33 sequence variants were deemed non-deleterious either because they were isocoding changes and would not be predicted to alter the encoded protein or were missense changes with an allele frequency > 4% and were previously described as polymorphisms in the NCBI SNP database (Table 3). One notable non-deleterious change was Leu1572Phe. It was found with a high frequency among USH2 patients but was not found among the patients with non-syndromic RP. It was determined that most patients with the Leu1572Phe change also had the common USH2 mutation c.2299delG; Glu767fs, and Leu1572Phe is thus likely over-represented in the USH2 group due to linkage disequilibrium with that mutation.

Table 3.

Isocoding changes and missense polymorphisms found in the coding region of exons 22–72 of the USH2A gene and their distribution among patients with USH2 or recessive non-syndromic retinitis pigmentosa.

distribution among all Usher type II
patients screened		 distribution among all ARRP patients
screened
exon nucleotide
change		 protein
change		 homoz hetero wild
type		 homoz hetero Wild
type
22 c.4714 C>T Leu1572Phe 2 38 170 0 0 145
25 c.4994T>C Ile1665Thr 2 24 68 3 18 52
25 c.5013C>A Gly1671Gly 5 35 54 4 16 53
27 c.5367A>G Leu1789Leu 0 1 165 0 0 140
27 c.5409C>G Val1803Val 0 1 165 0 1 139
28 c.2751C>T Tyr1917Tyr 0 1 216 0 0 149
31 c.6141 G>A Leu2047Leu 0 0 180 0 0 151
32 c.6186A>C Pro2062Pro 0 0 193 0 1 129
32 c.6270 G>A Leu2090Leu 0 1 192 0 0 130
32 c.6317C>T Thr2106Ile 22 68 103 15 55 60
33 c.6369 C>T Cys2123Cys 0 0 181 0 3 80
34 c.6506T>C Ile2169Thr 51 95 58 35 66 37
38 c.7131C>T Asn2377Asn 0 1 208 0 0 142
40 c.7506G>A Pro2502Pro 0 14 86 0 5 62
45 c.8937 A>G Val2979Val 0 0 106 0 1 79
46 c.9213 G>A Ser3071Ser 1 1 179 0 0 82
47 c.9296 A>G Asn3099Ser 0 6 99 0 12 64
48 c.9430G>A Asp3144Asn 0 6 99 0 11 69
52 c.10232 A>C Glu3411Ala 28 51 29 18 43 19
55 c.10836 C>A Val3612Val 0 0 99 0 1 72
59 c.11520T>C Tyr3840Tyr 0 1 80 0 0 53
60 c.11602A>G Met3868Val 11 25 71 3 22 55
61 c.11736G>A Glu3912Glu 0 0 186 0 2 143
61 c.11907 A>T Pro3969Pro 0 1 185 0 2 143
61 c.11946G>A Leu3982Leu 9 23 154 4 22 119
62 c.12093C>T Tyr4031Tyr 1 2 105 0 0 80
63 c.12597 T>C Ala4199Ala 0 0 204 0 1 140
63 c.12612A>G Thr4204Thr 27 56 121 12 38 91
63 c.12666A>G Thr4222Thr 21 97 86 23 55 63
63 c.13191G>A Glu4397Glu 6 24 77 6 17 56
63 c.13440 G>A Arg4480Arg 0 2 103 0 1 81
66 c.14481C>T Ala4827Ala 0 1 137 0 1 113
72 c.15522 T>C Tyr5174Tyr 0 0 107 0 2 76
Open in a new tab

The change Leu1572Phe was found to be in linkage disequilibrium with the common USH2 mutation c.2299delG; Glu767fs and is thus over-represented in the USH2 group. Homoz = number of patients homozygous for rare allele, Hetero = number of patients heterozygous for the rare allele, Wild-type = number of patients homozygous for the wild type, or normal sequence at the respective position. Total numbers are not equal as some samples did not yield data for some exons and some exons were screened in additional sets of patients (see methods).

Of the 108 patients originally diagnosed with USH2 who were included in the screen of the entire gene, 8 were found to carry deleterious mutations of the USH3A gene in a concurrent screen and were therefore re-categorized as USH3.[31] Combining the data from this study with the previous screen of exons 1–21, finds that 32 of the remaining 100 cases of USH2 have 2 clearly deleterious mutations and 25 have 1 clearly deleterious mutation of the USH2A gene. Therefore at least 57% of cases of USH2 carry at least one deleterious mutation of USH2A. If missense variations classified as either ‘likely deleterious” or “possibly deleterious” are in fact pathogenic mutations, then 37 patients would be counted as having 2 identified mutations and 26 with one mutation, for a total of 63% of patients having at least one detected USH2A mutation. If one assumes that patients with 2 mutations were all compound heterozygotes, then deleterious mutations were identified in 44.5% of the alleles screened. This number would increase to 50% of the alleles screened if all of the likely and possibly deleterious changes are included as mutations.

Among the 80 patients diagnosed with non-syndromic recessive retinitis pigmentosa, 3 had 2 clearly deleterious mutations and 12 had one deleterious mutation (19% total, including exons 1–21). By considering “likely deleterious” and “possibly deleterious” missense changes as mutations, the numbers change to 7 with 2 mutations and 11 with one mutation (23% total).

DISCUSSION

This study confirms earlier findings that mutations are distributed throughout the entire USH2A gene. The identification of 27 novel deleterious changes and 12 novel likely or possibly deleterious missense changes, each found in only one or a few patients, reinforces the idea that many different pathogenic USH2A mutations have arisen and most of which are confined to a few patients or small families. Our findings also corroborate earlier findings that Usher syndrome exhibits both nonallelic heterogeneity and extensive allelic heterogeneity. This intragenic heterogeneity presents difficulties for diagnostic screening of USH2A especially due to the large size of this gene. Efforts have been made recently to increase efficiency and reduce costs of mutation screening for known genes causing Usher syndrome by the development of genotyping microarrays using an arrayed primer extension method (APEX).[40] These arrays are only successful in identifying known mutations, and, since so many patients have unique mutations, they may fail to identify many USH2A mutations. These arrays can be used as a first quick screen of the gene but may require a follow-up detailed analysis of the entire gene in most patients.

We identified 65 rare missense variations most of which were novel to this set of patients. Functional analysis of such a large number of changes is not currently practical and analyses of normal controls will not provide statistically significant evidence for classifying many of them as pathogenic versus nonpathogenic. It is important to find an effective way to differentiate deleterious changes from benign ones. We developed a method of classifying these changes using the summed results of 4 different mutation assessment algorithms. This categorization system enabled us to tentatively define 8 of the 65 as “likely deleterious” and 9 as “possibly deleterious”. We must caution that the approach has not been validated for this gene and its specificity and sensitivity for categorizing missense changes are unknown. However recent studies of cancer genes have showed that when the output of several of these algorithms are in agreement the predictive value increases to 88.1% compared to 72.9%–82% when taken alone.[34] Even with combining the outcomes of 4 algorithms, the predictive value is not 100% as is exemplified by one of the changes identified in this study. The novel change, Phe2786Ser was classified as “likely deleterious” using our system, yet it was unlikely to be a cause of Usher syndrome since it was found in one patient who had 2 clearly pathogenic mutations elsewhere in the USH2A gene and another patient who had a mutation in the USH3A gene. Authors of some of these algorithms have cautioned that the predictions yielded by these programs should not be used for diagnostic purposes;[33] instead, the programs may help to select mutations that could be the focus of further analysis.

As with almost all screens for mutations in humans, many mutations are overlooked because the assays do not include most of the intron sequences, the 5’ and 3’ untranslated regions, the promoter region, and distant enhancer sequences. In addition, the assays do not detect large deletions, insertions, or rearrangements, and the sequencing of some exons of some samples fails for obscure technical reasons. Thus, our results can provide only an estimate of the prevalence of USH2A mutations in patients. With this in mind, the USH2A gene appears to be the major cause of USH2 with at least one mutation found in 57–63% of USH2 cases (percentages calculated without and including probable and possibly deleterious changes). Furthermore, USH2A causes a substantial number of cases of nonsyndromic retinitis pigmentosa (19–23% of cases). However since audiograms were not obtained, some of these patients may have un-reported hearing loss and may in fact, be USH2 patients. Given that the recessive form of RP accounts for 50–60% all cases of RP, that Usher syndrome represents 20–40% of the recessive cases,[41] and that Usher type II represents 50–75% of all Usher cases,[1, 3, 4] mutations in the USH2A gene likely account for 12 to 25% of all cases of RP (lower prevalence among USH2 {i.e. 0.57 × 0.5 × 0.2 × .5 = 2.9%} + lower prevalence among ARRP {i.e. 0.19 × 0.5 = 9.5%} combined = 12% to higher prevalence among USH2 {i.e. 0.63 × 0.6 × 0.4 × .75 = 11.3%} + higher prevalence among ARRP {i.e. 0.23 × 0.6 = 13.8%} combined = 25%). Since rhodopsin mutations account for 10% of cases of RP and RPGR account for 8% of cases, mutations in USH2A appear to be the most common known cause of RP in the United States.

ACKNOWLEDGEMENTS

Funding: This work was supported by the US National Institutes of Health grants NIH-EY00169, NIH-EY08683, and P30-EY014104, and The Foundation Fighting Blindness, Owings Mills, MD.

Footnotes

Competing interests: none

Copyright: “The Corresponding Author has the right to grant on behalf of all authors and does grant on behalf of all authors, an exclusive license (or non-exclusive for government employees) on a worldwide basis to the BMJ Publishing Group Ltd, and its Licensees to permit this article (if accepted) to be published in The Journal of Medical Genetics and any other BMJPGL products and to exploit all subsidiary rights, as set out in our license.”

REFERENCES

  • 1.Hope CI, Bundey S, Proops D, Fielder AR. Usher syndrome in the city of Birmingham--prevalence and clinical classification. Br J Ophthalmol. 1997;81:46–53. doi: 10.1136/bjo.81.1.46. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Boughman JA, Vernon M, Shaver KA. Usher syndrome: definition and estimate of prevalence from two high-risk populations. J Chronic Dis. 1983;36:595–603. doi: 10.1016/0021-9681(83)90147-9. [DOI] [PubMed] [Google Scholar]
  • 3.Rosenberg T, Haim M, Hauch AM, Parving A. The prevalence of Usher syndrome and other retinal dystrophy-hearing impairment associations. Clin Genet. 1997;51:314–321. doi: 10.1111/j.1399-0004.1997.tb02480.x. [DOI] [PubMed] [Google Scholar]
  • 4.Spandau UH, Rohrschneider K. Prevalence and geographical distribution of Usher syndrome in Germany. Graefes Arch Clin Exp Ophthalmol. 2002;240:495–498. doi: 10.1007/s00417-002-0485-8. [DOI] [PubMed] [Google Scholar]
  • 5.Sadeghi M, Kimberling W, Tranebjaerg L, Moller C. The prevalence of Usher Syndrome in Sweden: a nationwide epidemiological and clinical survey. Audiol Med. 2004;2:220–228. [Google Scholar]
  • 6.Weil D, Blanchard S, Kaplan J, Guilford P, Gibson F, Walsh J, Mburu P, Varela A, Levilliers J, Weston MD, et al. Defective myosin VIIA gene responsible for Usher syndrome type 1B. Nature. 1995;374:60–61. doi: 10.1038/374060a0. [DOI] [PubMed] [Google Scholar]
  • 7.Verpy E, Leibovici M, Zwaenepoel I, Liu XZ, Gal A, Salem N, Mansour A, Blanchard S, Kobayashi I, Keats BJ, Slim R, Petit C. A defect in harmonin, a PDZ domain-containing protein expressed in the inner ear sensory hair cells, underlies Usher syndrome type 1C. Nat Genet. 2000;26:51–55. doi: 10.1038/79171. [DOI] [PubMed] [Google Scholar]
  • 8.Bork JM, Peters LM, Riazuddin S, Bernstein SL, Ahmed ZM, Ness SL, Polomeno R, Ramesh A, Schloss M, Srisailpathy CR, Wayne S, Bellman S, Desmukh D, Ahmed Z, Khan SN, Kaloustian VM, Li XC, Lalwani A, Riazuddin S, Bitner-Glindzicz M, Nance WE, Liu XZ, Wistow G, Smith RJ, Griffith AJ, Wilcox ER, Friedman TB, Morell RJ. Usher syndrome 1D and nonsyndromic autosomal recessive deafness DFNB12 are caused by allelic mutations of the novel cadherin-like gene CDH23. Am J Hum Genet. 2001;68:26–37. doi: 10.1086/316954. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Bolz H, von BB, Ramirez A, Bryda EC, Kutsche K, Nothwang HG, Seeliger M, del C-S, Vila MC, Molina OP, Gal A, Kubisch C. Mutation of CDH23, encoding a new member of the cadherin gene family, causes Usher syndrome type 1D. Nat Genet. 2001;27:108–112. doi: 10.1038/83667. [DOI] [PubMed] [Google Scholar]
  • 10.Ahmed ZM, Riazuddin S, Bernstein SL, Ahmed Z, Khan S, Griffith AJ, Morell RJ, Friedman TB, Riazuddin S, Wilcox ER. Mutations of the protocadherin gene PCDH15 cause Usher syndrome type 1F. Am J Hum Genet. 2001;69:25–34. doi: 10.1086/321277. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Alagramam KN, Yuan H, Kuehn MH, Murcia CL, Wayne S, Srisailpathy CR, Lowry RB, Knaus R, Van LL, Bernier FP, Schwartz S, Lee C, Morton CC, Mullins RF, Ramesh A, Van CG, Hageman GS, Woychik RP, Smith RJ. Mutations in the novel protocadherin PCDH15 cause Usher syndrome type 1F. Hum Mol Genet. 2001;10:1709–1718. doi: 10.1093/hmg/10.16.1709. [DOI] [PubMed] [Google Scholar]
  • 12.Weil D, El-Amraoui A, Masmoudi S, Mustapha M, Kikkawa Y, Laine S, Delmaghani S, Adato A, Nadifi S, Zina ZB, Hamel C, Gal A, Ayadi H, Yonekawa H, Petit C. Usher syndrome type I G (USH1G) is caused by mutations in the gene encoding SANS, a protein that associates with the USH1C protein, harmonin. Hum Mol Genet. 2003;12:463–471. doi: 10.1093/hmg/ddg051. [DOI] [PubMed] [Google Scholar]
  • 13.Chaib H, Kaplan J, Gerber S, Vincent C, Ayadi H, Slim R, Munnich A, Weissenbach J, Petit C. A newly identified locus for Usher syndrome type I, USH1E, maps to chromosome 21q21. Hum Mol Genet. 1997;6:27–31. doi: 10.1093/hmg/6.1.27. [DOI] [PubMed] [Google Scholar]
  • 14.Eudy JD, Weston MD, Yao S, Hoover DM, Rehm HL, Ma-Edmonds M, Yan D, Ahmad I, Cheng JJ, Ayuso C, Cremers C, Davenport S, Moller C, Talmadge CB, Beisel KW, Tamayo M, Morton CC, Swaroop A, Kimberling WJ, Sumegi J. Mutation of a gene encoding a protein with extracellular matrix motifs in Usher syndrome type IIa. Science. 1998;280:1753–1757. doi: 10.1126/science.280.5370.1753. [DOI] [PubMed] [Google Scholar]
  • 15.Weston MD, Luijendijk MW, Humphrey KD, Moller C, Kimberling WJ. Mutations in the VLGR1 gene implicate G-protein signaling in the pathogenesis of Usher syndrome type II. Am J Hum Genet. 2004;74:357–366. doi: 10.1086/381685. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Ebermann I, Scholl HP, Charbel IP, Becirovic E, Lamprecht J, Jurklies B, Millan JM, Aller E, Mitter D, Bolz H. A novel gene for Usher syndrome type 2: mutations in the long isoform of whirlin are associated with retinitis pigmentosa and sensorineural hearing loss. Hum Genet. 2007;121:203–211. doi: 10.1007/s00439-006-0304-0. [DOI] [PubMed] [Google Scholar]
  • 17.Ben RI, Benzina Z, Dhouib H, Hadjamor I, Amyere M, Ayadi L, Turki K, Hammami B, Kmiha N, Kammoun H, Hakim B, Charfedine I, Vikkula M, Ghorbel A, Ayadi H, Masmoudi S. Identification of candidate regions for a novel Usher syndrome type II locus. Mol Vis. 2008;14:1719–1726. [PMC free article] [PubMed] [Google Scholar]
  • 18.Joensuu T, Hamalainen R, Yuan B, Johnson C, Tegelberg S, Gasparini P, Zelante L, Pirvola U, Pakarinen L, Lehesjoki AE, de la CA, Sankila EM. Mutations in a novel gene with transmembrane domains underlie Usher syndrome type 3. Am J Hum Genet. 2001;69:673–684. doi: 10.1086/323610. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Ness SL, Ben-Yosef T, Bar-Lev A, Madeo AC, Brewer CC, Avraham KB, Kornreich R, Desnick RJ, Willner JP, Friedman TB, Griffith AJ. Genetic homogeneity and phenotypic variability among Ashkenazi Jews with Usher syndrome type III. J Med Genet. 2003;40:767–772. doi: 10.1136/jmg.40.10.767. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Pakarinen L, Karjalainen S, Simola KO, Laippala P, Kaitalo H. Usher's syndrome type 3 in Finland. Laryngoscope. 1995;105:613–617. doi: 10.1288/00005537-199506000-00010. [DOI] [PubMed] [Google Scholar]
  • 21.Baux D, Larrieu L, Blanchet C, Hamel C, Ben SS, Vielle A, Gilbert-Dussardier B, Holder M, Calvas P, Philip N, Edery P, Bonneau D, Claustres M, Malcolm S, Roux AF. Molecular and in silico analyses of the full-length isoform of usherin identify new pathogenic alleles in Usher type II patients. Hum Mutat. 2007;28:781–789. doi: 10.1002/humu.20513. [DOI] [PubMed] [Google Scholar]
  • 22.Aller E, Jaijo T, Beneyto M, Najera C, Oltra S, Ayuso C, Baiget M, Carballo M, Antinolo G, Valverde D, Moreno F, Vilela C, Collado D, Perez-Garrigues H, Navea A, Millan JM. Identification of 14 novel mutations in the long isoform of USH2A in Spanish patients with Usher syndrome type II. J Med Genet. 2006;43:e55. doi: 10.1136/jmg.2006.041764. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Dreyer B, Brox V, Tranebjaerg L, Rosenberg T, Sadeghi AM, Moller C, Nilssen O. Spectrum of USH2A mutations in Scandinavian patients with Usher syndrome type II. Hum Mutat. 2008;29:451. doi: 10.1002/humu.9524. [DOI] [PubMed] [Google Scholar]
  • 24.Weston MD, Eudy JD, Fujita S, Yao S, Usami S, Cremers C, Greenberg J, Ramesar R, Martini A, Moller C, Smith RJ, Sumegi J, Kimberling WJ. Genomic structure and identification of novel mutations in usherin, the gene responsible for Usher syndrome type IIa. Am J Hum Genet. 2000;66:1199–1210. doi: 10.1086/302855. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.van WE, Pennings RJ, te BH, Claassen A, Yntema HG, Hoefsloot LH, Cremers FP, Cremers CW, Kremer H. Identification of 51 novel exons of the Usher syndrome type 2A (USH2A) gene that encode multiple conserved functional domains and that are mutated in patients with Usher syndrome type II. Am J Hum Genet. 2004;74:738–744. doi: 10.1086/383096. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Baux D, Faugere V, Larrieu L, Le Guedard-Mereuze S, Hamroun D, Beroud C, Malcolm S, Claustres M, Roux AF. UMD-USHbases: a comprehensive set of databases to record and analyse pathogenic mutations and unclassified variants in seven Usher syndrome causing genes. Hum Mutat. 2008;29:E76–E87. doi: 10.1002/humu.20780. [DOI] [PubMed] [Google Scholar]
  • 27.Dreyer B, Tranebjaerg L, Brox V, Rosenberg T, Moller C, Beneyto M, Weston MD, Kimberling WJ, Cremers CW, Liu XZ, Nilssen O. A common ancestral origin of the frequent and widespread 2299delG USH2A mutation. Am J Hum Genet. 2001;69:228–234. doi: 10.1086/321269. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Tavtigian SV, Greenblatt MS, Lesueur F, Byrnes GB. In silico analysis of missense substitutions using sequence-alignment based methods. Hum Mutat. 2008;29:1327–1336. doi: 10.1002/humu.20892. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Seyedahmadi BJ, Rivolta C, Keene JA, Berson EL, Dryja TP. Comprehensive screening of the USH2A gene in Usher syndrome type II and non-syndromic recessive retinitis pigmentosa. Exp Eye Res. 2004;79:167–173. doi: 10.1016/j.exer.2004.03.005. [DOI] [PubMed] [Google Scholar]
  • 30.Sandberg MA, Rosner B, Weigel-DiFranco C, McGee TL, Dryja TP, Berson E. Disease Course in Patients with Autosomal Recessive Retinitis Pigmentosa due to the USH2A Gene. Invest Ophthalmol Vis Sci. 2008 doi: 10.1167/iovs.08-2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Seyedahmadi BJ, Dryja TP, Berson EL. USH3A mutations in patients with a prior diagnosis of Usher syndrome type I, Usher syndrome type II, and nonsyndromic recessive retinitis pigmentosa. Invest Ophthalmol Vis Sci. 2004;45 E-abstract 4726. [Google Scholar]
  • 32.Ng PC, Henikoff S. SIFT: Predicting amino acid changes that affect protein function. Nucleic Acids Res. 2003;31:3812–3814. doi: 10.1093/nar/gkg509. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Kumar P, Henikoff S, Ng PC. Predicting the effects of coding non-synonymous variants on protein function using the SIFT algorithm. Nat Protoc. 2009;4:1073–1081. doi: 10.1038/nprot.2009.86. [DOI] [PubMed] [Google Scholar]
  • 34.Chan PA, Duraisamy S, Miller PJ, Newell JA, McBride C, Bond JP, Raevaara T, Ollila S, Nystrom M, Grimm AJ, Christodoulou J, Oetting WS, Greenblatt MS. Interpreting missense variants: comparing computational methods in human disease genes CDKN2A, MLH1, MSH2, MECP2, and tyrosinase (TYR) Hum Mutat. 2007;28:683–693. doi: 10.1002/humu.20492. [DOI] [PubMed] [Google Scholar]
  • 35.Ramensky V, Bork P, Sunyaev S. Human non-synonymous SNPs: server and survey. Nucleic Acids Res. 2002;30:3894–3900. doi: 10.1093/nar/gkf493. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Ferrer-Costa C, Gelpi JL, Zamakola L, Parraga I, de IC X, Orozco M. PMUT: a web-based tool for the annotation of pathological mutations on proteins. Bioinformatics. 2005;21:3176–3178. doi: 10.1093/bioinformatics/bti486. [DOI] [PubMed] [Google Scholar]
  • 37.Tavtigian SV, Deffenbaugh AM, Yin L, Judkins T, Scholl T, Samollow PB, De SD, Zharkikh A, Thomas A. Comprehensive statistical study of 452 BRCA1 missense substitutions with classification of eight recurrent substitutions as neutral. J Med Genet. 2006;43:295–305. doi: 10.1136/jmg.2005.033878. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Yan D, Ouyang X, Patterson DM, Du LL, Jacobson SG, Liu XZ. Mutation analysis in the long isoform of USH2A in American patients with Usher Syndrome type II. J Hum Genet. 2009 doi: 10.1038/jhg.2009.107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Auslender N, Bandah D, Rizel L, Behar DM, Shohat M, Banin E, lon-Shalev S, Sharony R, Sharon D, Ben-Yosef T. Four USH2A founder mutations underlie the majority of Usher syndrome type 2 cases among non-Ashkenazi Jews. Genet Test. 2008;12:289–294. doi: 10.1089/gte.2007.0107. [DOI] [PubMed] [Google Scholar]
  • 40.Cremers FP, Kimberling WJ, Kulm M, de Brouwer AP, van WE, te BH, Cremers CW, Hoefsloot LH, Banfi S, Simonelli F, Fleischhauer JC, Berger W, Kelley PM, Haralambous E, Bitner-Glindzicz M, Webster AR, Saihan Z, De BE, Leroy BP, Silvestri G, McKay GJ, Koenekoop RK, Millan JM, Rosenberg T, Joensuu T, Sankila EM, Weil D, Weston MD, Wissinger B, Kremer H. Development of a genotyping microarray for Usher syndrome. J Med Genet. 2007;44:153–160. doi: 10.1136/jmg.2006.044784. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Hartong DT, Berson EL, Dryja TP. Retinitis pigmentosa. Lancet. 2006;368:1795–1809. doi: 10.1016/S0140-6736(06)69740-7. [DOI] [PubMed] [Google Scholar]

RESOURCES